Virtual surgical systems and methods

Abstract

Various embodiments of virtual surgical systems and methods are disclosed.

1. A system for virtually reconstructing a fractured bone, comprising:
 a memory with software stored therein; and 
 a processor configured with the software to:
 match fracture surfaces of an imaged fractured bone with a data aligned rigidity constrained exhaustive search (DARCES) surface matching algorithm that uses fracture surface data sets to produce a transformed sample data set for at least one of the fracture surfaces; 
 match the fracture surfaces with an iterative closest point (ICP) algorithm that uses the transformed sample data set; and 
 minimize a mean-squared error between the matched fracture surfaces based on iterative angular perturbations until a desired reconstruction of the fractured bone is achieved; and 
 wherein the processor is further configured with the software to: 
 receive image slices corresponding to the imaged bone having a hairline fracture, the image slices each comprising a plurality of pixel blocks; 
 fit quadratic polynomial functions to inner and outer contours of the imaged bone for select pixel blocks less than the entire plurality of the pixel blocks for each of the images slices; 
 iteratively sample the select pixel blocks based on Markov random field modeling and maximum a posteriori estimation to visualize the hairline fracture; and 
 reconstruct the bone at the hairline fracture site in a manner to mimic the natural bone healing process in the absence of any surgical intervention. 
 
 
     
     
       2. The system of  claim 1 , wherein the processor is further configured with the software to:
 receive data corresponding to image slices of the imaged bone comprising a fracture; 
 determine potential fracture points of the imaged bone from the data based on a curvature scale space algorithm; 
 filter out one or more of the potential fracture points; and 
 determine which of the remaining fracture points are stable across a predetermined number of the image slices based on a Kalman filter that provides surface data for the fracture. 
 
     
     
       3. The system of  claim 1 , wherein the processor is further configured with the software to:
 identify at least two fracture surface pairs based on a maximum weight graph matching algorithm; 
 model the fracture surfaces as vertices of a general graph; 
 compile a score matrix comprising edge weights between the vertices of the fracture surface pairs to determine opposing fracture surface pairs; and 
 register the opposing fracture surface pairs.